Hi all,
I am using an AD5254 (1k version) and an arduino to simulate a pt1000 sensor.
Actually I have a device that use pt1000 for measure temperature and instead plug in the pt1000 sensor I wired my arduino and digital pot like in the attached picture.
I am changing the W0 value every >50ms from arduino.
The issue I have is that at some point (i haven t been able to find a pattern) the AD5254 getting very very hot (2 of them just had smoke out).
Also when is getting hot it doesn't change resistance anymore.
What I am doing wrong?!

Some important info:
Voltage measured without the pt1000 connected (on the device that actually read pt1000) is about 13.3V.
Voltage measured with AD5254 connected like in the picture is around 1.5V.
AD5254 is powered and controlled from arduino mega.

Any idea of what I can do for making work cold?

 

Do you have the bypass capacitors suggested in the datasheet? Should AD0 be pinned to Vss or Vdd, as AD1 is?

A clear photo of your build might help someone identify a problem.

 

Hi,
Ad0, vss and ad1 are wired to gnd.
I don t have capacitors instead....

 

Voltage measured without the pt1000 connected (on the device that actually read pt1000) is about 13.3V.

Don't know exactly what that means but 13.3 volts doesn't sound good.
What is connected after the two resistor R4 and R2 where you show the two arrows?
SG

 

Don't know exactly what that means but 13.3 volts doesn't sound good.
What is connected after the two resistor R4 and R2 where you show the two arrows?
SG

The device that read pt1000. I either connect ad5254 either pt1000.
13 3V i have with nothing connected to the 2 arrows...when i connect my pot or pt1000 sensor i have around 1.5V...so I don t havr 13V on the digital pot...

 

Where is the 13.3 volts coming from ?
SG

 

Hi,
Ad0, vss and ad1 are wired to gnd.
I don t have capacitors instead....

Installed? At the very least you should put a 0.1 microfarad ceramic capacitor across the power pins of that IC. Oscillation is a common source of overheating.

 

I will put now the capacitors but other question came into my mind: Is that ok I left "B1" in the air? maybe I should ground it or something?

I am looking on fig 40 from 40 and there are also capacitors between Vss and GND. In my case since Vss is wired to GND I will punt only the 2 capacitors between Vdd and GND

 

I redo all the wiring and just play with it like in pictures attached:
I put the 2 capacitors on the power to AD5254.
I put together W0 and B0.

Findings :
If I measure voltage between grey wires W0/A0 i have 0. Is that normal?
I also measured resistance and is moving once I wrote values so it acting normal.
At some point I saw that was again overheated and I unplug it.
Then put back again and till now is cold and is working normal?
What can cause that overheating in that case? Only multimetter was tied to grey wires....I can t reproduced it now.

 

If I measure voltage between grey wires W0/A0 i have 0. Is that normal?

If you are measuring voltage then there must be a voltage source. The photo of the Arduino is no help, where is the photo of board with the AD5254 with wires attached.
The schematic below shows one way the digital pot can be wired to simulate a pt1000.
I think the problem with overheating is you are pushing too much current through the pot.
SG

 

Max current measured was 1.2 mA. Actually i had between 1 and 1.2 mA when simulating 17 to 99 Celsius degrees.
....is still on testing.
Today i let it working around 3 hours simulating different temperatures and was cold all the time...

 

Max current measured was 1.2 mA. Actually i had between 1 and 1.2 mA when simulating 17 to 99 Celsius degrees.
....is still on testing.
Today i let it working around 3 hours simulating different temperatures and was cold all the time...

Great maybe the capacitors did the trick.
SG

 

Great maybe the capacitors did trick

Beside add capacitors, other mod i did was to tie together b0 and w0 ...not sure how relevant is this

 

Beside add capacitors, other mod i did was to tie together b0 and w0 ...not sure how relevant is this

Should not be a problem as long as the current does not exceed 5ma
SG

 

If you are measuring voltage then there must be a voltage source. The photo of the Arduino is no help, where is the photo of board with the AD5254 with wires attached.
The schematic below shows one way the digital pot can be wired to simulate a pt1000.
I think the problem with overheating is you are pushing too much current through the pot.
SG
View attachment 171125

Sorry, i haven t noticed that 2 picture missing from the attachment
I will put them now again
Also I found this on datasheet
"
POWER-UP AND POWER-DOWN SEQUENCES
Because the ESD protection diodes limit the voltage compliance
at Terminal A, Terminal B, and Terminal W (Figure 39), it is important to power VDD/VSS before applying any voltage to these terminals. Otherwise, the diodes are forward biased such that VDD/VSS are powered unintentionally and may affect the
user’s circuit. Similarly, VDD/VSS should be powered down last.
The ideal power-up sequence is in the following order: GND, VDD, VSS, digital inputs, and VA/VB/VW. The order of powering VA, VB, VW, and the digital inputs is not important, as long as
they are powered after VDD/VSS."

That means I should add some relays for connection to the device that measure pt1000 because it is always connected to my simulator but my simulator is not always ON. So I am thinking to connect the device through relays and when my arduino is of, there is no connection between device and my ad5254.

Thanks

 

I should add some relays for connection to the device that measure pt1000

Relay or a transistor switching.
OK from your PC board I see you are using two digital pots. What is connected to the pots using the gray wires and the pair of yellow wires?
SG

 

Hi, the grey ones I am using for simulating pt1000
The yellow ones are not used..i put there for testing purpose